Family size has long been a barometer of the current and future prospects of parents and their children, but the implications of living in a large or small family for children have evolved considerably over time and across societies. When extended family ties are strong and children can provide financial resources to the household, a large family may not necessarily imply disadvantages for all children. With economic development and increasing returns to education, the benefits parents accrue from having many children become increasingly at odds with the educational disadvantages faced by adolescents with many siblings. This tension is vividly seen in rapidly emerging economies such as Brazil, and highlights the challenging educational gaps faced by such countries. This project, therefore, will examine the links between family size and adolescent education in Brazil, providing a valuable contrast to the U.S.-based literature on this topic, and extending recent research that challenges the long-standing conclusion that there is a direct tradeoff between having a large family and investing in children's schooling. While the early empirical literature has generally confirmed the theoretical prediction that family size is negatively related to children's education, a recent stream of research has cast doubt on the homogeneity of these findings on conceptual and methodological grounds. Because of the incredible temporal and regional variation in social conditions, fertility regimes, and education Brazil has experienced over the last decades, the country offers an excellent opportunity to sort out the conflicting evidence on how family size predicts children's education. Brazil also offers high-quality nationally representative data that cover most of the years of the demographic transition as well as the dramatic changes in socio-economic conditions the country has experienced. This project extends previous research on the connections between family size and children's education by examining historical, regional, and gender variation in the case of Brazil, while at the same time addressing methodological concerns about the joint determination of education and family size that have recently gained traction in the literature. To accomplish this goal, we will implement twins and same-sex siblings' instrumental variable approaches using 32 years of nationally representative data, the PNAD (1977- 2009). This research is important because understanding whether and how the effect of family size on adolescent education varies by social, demographic and economic context is the first step toward addressing broader inequalities in social opportunities, a concern of particular relevance for Brazil, one of the most unequal countries in the world. The public health relevance of this project also comes from well-established evidence that rising children's educational levels improve social capacity for population health with substantial payoffs both in terms of reducing the burden of health problems at the societal level and in improving the quality of life of individuals.